Exam 1 Flashcards
Theory
Explanation based on observations
Hypothesis
Prediction based on theory
Research
Test of the hypothesis, yields data that either supports or refutes
Aspects of a good theory
-falsifiable
-motivates new hypotheses
-parsimonious
Replication
Repetition of a research study to confirm the results
Reasons a study may fail to replicate
-benign: differences in study design across replications
-less benign: original study was a false positive
-even less benign: results of the original study were doctored to support the original hypothesis
Independent variable
Researcher actively manipulates one or more variables
Dependent variable
Measures how the variables affect the participants’ responses
Independent-groups (between-subject) design
-Each subject participates in only one condition of the IV
-Random assignment
Dependent-groups (within-subject) design
-Every subject participates in all conditions of the IV
-Counterbalancing: Order of conditions is varied across subjects
Descriptive Research
-Researcher measures variables but does not actively manipulate
them
-Results describe characteristics of variables and relationships
between variables
-Does not provide conclusions about cause
Types of descriptive research
-Case studies
-Observational studies
-Surveys
-Correlation studies
Correlation
A measure of the relationship between two variables
Correlation coefficient
A mathematical estimate of the strength of this relationship
Scatter plot
A convenient way to show correlation graphically
Correlation does not
Establish causation
Quantitative Research
Relies primarily on numerical data and statistical analysis to describe
and understand human behavior
Qualitative Research
-Seeks to achieve thematic description and understanding of behavior,
primarily through nonstatistical analysis of data
-May examine data without the use of statistics
Mixed-methods research
Leverages a combination of quantitative and qualitative methods
Laboratory research
-Lab setting affords most control over experimental factors
-The effects of the IV can be isolated
-High internal validity – We can be confident in the results of the study
Field research
-Research is conducted in a more realistic setting
-High external validity – The results likely apply to a broader context
Population
The group that a researcher
is interested in examining defined by specific characteristics such as residency, occupation, gender or age
Convenience Sample
Chosen for ease
of study (e.g., undergraduate studies,
polling in a single city or county). Results probably do not generalize to the population
Representative sample
Resembles the population (e.g., percent male/female,
percent young/old, demographics)
Random sample
Every member of the
population has an equal chance of being drawn (more well-suited to represent the population)
Random sampling
Uses random selection (all members of a particular population or subpopulation have an equal chance of being selected)
Random sampling with replacement:
A selected member is returned to the pool and may be selected again.
Random sampling without replacement:
Once selected, a member is removed from the pool
and cannot be selected again
Stratified random sampling:
Sampling key subpopulations based on
specific characteristics in the same proportion as in the population
Convenience sampling:
Consists of those who are available and willing to
participate
Quota sampling:
Sample represents key subpopulations
Maximum variation sampling:
Sample represents the full range of extremes
Internal replication
Occurs when researchers follow up their initial study with one or more replications and present this series of studies in a single research report.
Independent replication
Is a replication conducted by researchers who were not part of the original research group.
Complete (full) replications
Precisely mirror the original study.
Partial replications
Include some aspects of the original study.
Direct (exact) replication
Researchers mimic original procedures.
Conceptual replications
Examine the same research question but operationalize constructs differently.
Replication with extension
Add a new design element to the original study.
Construct validity
Are we studying what we intend to study?
Statistical validity
How thorough are the statistics that we used to back up our findings
Internal validity
How much confidence can we have in our
experimental result
External validity
How generalizable are our findings?
Seven basic threats to internal validity
- History
- Maturation
- Testing
- Instrumentation
- Regression to the mean
- Attrition (subject loss)
- Selection
History
Events to which people are exposed while participating in a study, but that are not part of the experimental manipulation
-Solution(s): Block randomization in dependent-groups studies
Maturation
Ways in which people naturally change over time, independent of their participation in a study
-Solution: Random assignment to experimental groups
Testing
The act of measuring individuals’ responses may affect their responses
on subsequent measures.
-Solution: Avoid pretesting or ensure that all participants complete a pretest.
Instrumentation
Changes that occur in a measuring instrument during data collection
-Solution: Random assignment; exclude participants with extreme scores
Regression to the mean
The statistical concept that when two variables are not perfectly correlated, more extreme scores on one variable will be associated
overall with less extreme scores on the other variable
-Solution: Random assignment; exclude participants with extreme scores
Attrition (subject loss)
Participants fail to complete a study
-Solution:
Establish why participants drop out
Establish whether participants who remain in differ from participants
who left
Selection
At the start of a study, participants in the various conditions already
differ on a characteristic that can partly or fully account for the eventual
results
-Solution(s): Random assignment
Descriptive statistics
The numbers used to summarize the characteristics of a sample.
Frequency
Count of how many times a score occurred
Percentage
Proportion of a score within a sample
Cumulative percentage:
Proportion of the sample that falls within a specified interval
Positively skewed:
Tail is on the right with more scores clustered at the low end of the scale
Negatively skewed:
Tail is on the left with more scores clustered at the high end of the scale
Uniform Distribution
All scores have roughly the same frequency
Measures of central tendency
Median, mean, mode
Median
The score that cuts the sample in half; 50th percentile
Mean
The arithmetic average
Mode
The most frequent score in a sample
Range:
Distance between the minimum and maximum score
Standard deviation:
How much, in general, scores differ from the mean
Variance:
Standard deviation squared
Inferential statistics:
Statistical analysis of data gathered from a sample to draw conclusions about a population from which the
sample is drawn.
Hypothesis testing:
The process of determining the probability of obtaining a particular result or set of results
Sampling distribution:
A distribution obtained from multiple samples of the same size drawn from the same population
Null hypothesis (Ho):
A prediction of no difference between groups of the dependent variable
Alternative hypothesis (Ha):
A prediction of the result if the null hypothesis is
not true
Criterion of p < .05
Indicates that there is less than a 5% chance that the
results were due to chance alone.
P = 0.05 means
If the null hypothesis is true, then there was a 5% chance that we would obtain our result. There is a 5% chance that our result was due entirely to chance
Type I error:
The probability of rejecting a true Ho; defined by the probability of the significance level of your findings
“false positive”
Type II error:
The probability of incorrectly retaining a false Ho
“false negative”